hardware multiply/divide functionality in CPUs

[Brent Hilpert]

On 2011 Mar 7, at 4:08 PM, Jules Richardson wrote:
> Richard wrote:
>> I don't think the multiply implementation in the IBM 701 was shift/add
>> microcode variety.  That wouldn't have been very performant for its
>> intended purpose (scientific computing).
>
> I'll have to have a look online and see if I can find any details. I 
> see that Booth's algorithm was 1951 - much earlier than I expected and 
> a year before the 701. I'm not sure when Wallace trees showed up...

I know you mentioned 70s/80s in your initial message, but if you are 
considering early stuff:

- The ENIAC did hardware multiple/divide/square root. Multiply was done 
using a partial-products scheme.

- Some IBM card/unit-record/accounting machines did electronic 
multiply/divide in the 1940s/50s. I sometimes see mention of an IBM 
researcher developing an electronic multiplier in the late-30s or very 
early-40s but I don't have a ref.

- The Draft Report on the EDVAC (1945) described or included hardware 
multiply and divide. I don't know whether it was implemented on the 
EDVAC or not.

- Whirlwind had hardware multiply (not sure about divide), late-40s.

I wonder if multiply/divide were the incentive for Wilkes' development 
of microcode, as most other instructions of the time would not be very 
complex.